Return-Path: From: Bing Zhao To: Marcel Holtmann CC: Linux Bluetooth mailing list , "Gustavo F. Padovan" , Johan Hedberg , Amitkumar Karwar , Avinash Patil , Chin-Ran Lo , Xinming Hu Date: Fri, 18 Jul 2014 14:46:26 -0700 Subject: RE: [PATCH 2/2] Bluetooth: btmrvl: add public address configuration support Message-ID: <477F20668A386D41ADCC57781B1F7043122FE2C674@SC-VEXCH1.marvell.com> References: <1405538486-23533-1-git-send-email-bzhao@marvell.com> <1405538486-23533-2-git-send-email-bzhao@marvell.com> <477F20668A386D41ADCC57781B1F7043122FCE9C75@SC-VEXCH1.marvell.com> In-Reply-To: <477F20668A386D41ADCC57781B1F7043122FCE9C75@SC-VEXCH1.marvell.com> Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 List-ID: Hi Marcel, > > > + ret =3D btmrvl_send_sync_cmd(priv, BT_CMD_SET_BDADDR, buf, sizeof(b= uf)); > > > > So I keep wondering why you need the btmrvl_send_sync_cmd here. The HCI= transport is up and running > > at this stage and sending vendor command should do the right thing. Sim= ilar to what you do in btusb > > driver. > > > > You might have explained this to, but I must have clearly not understoo= d it. >=20 > We will check if btmrvl_send_sync_cmd() can be replaced by __hci_cmd_sync= () and get back to you. __hci_cmd_sync() sets skb->packet_type as 0x01 whereas btmrvl_send_sync_cmd= () sets it as 0xfe. But the firmware handles both 0x01 and 0xfe tx packets in similar way. Ther= efore we can replace btmrvl_send_sync_cmd() with __hci_cmd_sync(). I will send a v2 for this patch shortly. Later on we will send out a cleanu= p patch for other existing commands that are using btmrvl_send_sync_cmd(). Thanks, Bing